chore: use io.ReadFull instead of io.ReadAll for DataAsync (#12795)

Since the final size is already known, no need to `ReadAll` a `LimitedReader`: directly `ReadFull` a properly sized buffer.

Tests are already present in `blob_test.go` (a failure can be triggered by creating a smaller `buf`).

`go test -run=TestBlob_Data -bench=Blob_Data -benchmem` before:
```
Benchmark_Blob_Data-18             43964             28727 ns/op            1373 B/op         11 allocs/op
```
After:
```
Benchmark_Blob_Data-18             41308             27679 ns/op             846 B/op         10 allocs/op
```

🎉 one allocation spared!

Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/12795
Reviewed-by: Gusted <gusted@noreply.codeberg.org>

modules/git/blob.go

@@ -83,13 +83,14 @@ func (b *Blob) DataAsync() (io.ReadCloser, error) {
 	}
 
 	if size < 4096 {
-		bs, err := io.ReadAll(io.LimitReader(rd, size))
+		buf := make([]byte, size)
+		_, err := io.ReadFull(rd, buf)
 		defer cancel()
 		if err != nil {
 			return nil, err
 		}
 		_, err = rd.Discard(1)
-		return io.NopCloser(bytes.NewReader(bs)), err
+		return io.NopCloser(bytes.NewReader(buf)), err
 	}
 
 	return &blobReader{

modules/git/blob_test.go

@@ -143,7 +143,7 @@ func Benchmark_Blob_Data(b *testing.B) {
 		b.Fatal(err)
 	}
 
-	for i := 0; i < b.N; i++ {
+	for b.Loop() {
 		r, err := testBlob.DataAsync()
 		if err != nil {
 			b.Fatal(err)